1
Promoting End-of-life Advanced Care
Planning using Health IT
Session # 239, February 14, 2019
Jonathan Austrian MD, Medical Director, Inpatient Clinical Informatics
Glenn Doty RN, Senior Director, Clinical Systems & Transformation
2
Guidance Outside Health System
3
“Two interventions have consistently been shown to
help patients live their final days in accordance with
their wishes: earlier conversations about their goals
and greater use of palliative care services…”
- New York Times (May 10, 2017)
Guidance Outside Health System
4
Supportive Care Program
Awareness of Patient Preferences
Inpatient Supportive Care Protocols
Screening
Data/Analytics
Change Management
5
Jonathan Austrian MD
Has no real or apparent conflicts of interest to report.
Glenn Doty RN
Has no real or apparent conflicts of interest to report.
Conflict of Interest
6
• Background
• Case for Change
• Interventions
• Outcomes
• Barriers
Agenda
7
• Discuss the rationale for investing in Health IT interventions for
goals of care
• Diagram the clinical workflow from screening to intervention for
patients who will benefit from goals of care interventions
• Design Clinical Decision Support Interventions to promote
screening for patients who could benefit from Goals of Care
Conversations
• Identify barriers to adoption of goals of care health IT interventions
• Evaluate the impact of goals of care health IT interventions on
important clinical care process and outcome measures
Learning Objectives
8
• Health system based in New York City with locations across
the five boroughs, Westchester, Putnam and Dutchess
Counties, New Jersey, Long Island and Florida
• 230 locations including 6 inpatient facilities
• 3,600+ physicians serving over 3 million patients a year
• #3 best medical school for research and #15 best hospital in
the US
• Among 9 percent of hospitals nationwide to earn a 5-star
rating for safety, quality, and patient experience from the
Centers for Medicare and Medicaid Services
• Winner of the 2018 HIMSS Davies Award for demonstrating
outstanding achievement in utilizing health information
technology to substantially improve patient outcomes and
value
NYU Langone Health
10
Key Findings:
•12% of discharges and 18.6% of the
average daily census were end-of-life (EOL)
patients
•26% of PICC lines, 42% of PEG tubes, and
38% of Tracheostomies were placed on
EOL patients
•EOL patients compared to entire population:
•2.05x Readmission Rate
•1.8x Infection Rate
•+2.9 days greater ICU length of stay
Quality and Utilization Analysis
“I would not be
surprised if this
patient passed away
in the next 6
months”
Hospice
Expired
+
+
End of Life (EOL)
Cohort:
Compared one year of adult
inpatient activity for EOL cohort
against entire patient population
11
Supportive Care Pillars
Awareness of Patient Preferences
Inpatient Supportive Care Protocols
Screening
Data/Analytics
Change Management
12
Supportive Care Pillars
Change Management
13
The Mission of NYU’s Supportive
Care Program
• To improve the quality of life for our end of life patients
• Better align our clinical practice with the patient’s stated
goals
• Empower our providers to give stronger guidance to
patients and families on what is appropriate at the end
of life
• To reduce non-value added inpatient utilization in
patients near the end of life
14
Supportive Care Program Goals
#
Goal Description
Quality
1
Reduce
readmission rate for end-of-life (EOL) patients in the
last 6 months of life
AND reduce proportion of all readmissions
that are incurred by EOL patients
2
Reduce
total number of hospital-acquired conditions (HACs)
in
the last 6 months of life for
EOL patients
3
Reduce
ED Visits for Oncology patients in last 30 days of life
Cost
4
Reduce
Total Patient Days
for EOL patients in the last 6 months
and 30 days of life
5
Reduce overall inpatient
variable direct cost in last 6 months of
life for EOL
patients
EOL Cohort defined as adult patients that were discharged to hospice or expired
15
Clinical Workflow
16
Supportive Care Pillars
Awareness of Patient Preferences
17
1
Advance Care Planning Navigator in
Epic:
18
1
8
Health Care
Agents
Patient Capacity
ACP Activation
Note
ACP History
19
1
9
Health Care Agents
Patient
Capacity
ACP Activation
Note
ACP History
Patient
Capacity
HCA
Patient
Header
Full capacity
None
Incapacitated
(HCA indicated)
Active
Incapacitated
(No
HCA
indicated)
Not on file
Needs Review
Pending
20
2
0
Health Care AgentsPatient Capacity
ACP Activation
Note
ACP History
21
2
1
• eMOLST
• Code Status History
• Prior ACP Notes
• Prior ACP Documents
Health Care AgentsPatient Capacity
ACP
NoteActivation
ACP History
22
Supportive Care Pillars
Screening
23
Based on their clinical expertise, our providers answered
the following question for admitted patients:
This is called the “Mandatory Surprise Question” or “MSQ”
How Did We Identify Patients That
Would Benefit?
Would you be surprised if this patient passed
away in the next 6 months?
24
The Mandatory Surprise Question
(MSQ)
THE MSQ IS AN OPPORTUNITY FOR THE PROVIDER
TO:
1. Quickly identify patients that may be near the end of
life
2.
Pause to consider possible modifications to the course
of treatment
3.
Make key decisions about the patient’s care trajectory
that are in line with Supportive Care best practices
25
MSQ
26
Mortality Predictive Analytics
27
Supportive Care Pillars
Inpatient Supportive Care Protocols
28
Standards
• All patients Screened with MSQ
• All MSQ = No patients should have ACP Note
• All Predictive Analytics patients should have ACP Note
• All DNR patients should have eMolst
29
Standard Documentation ACP note
30
Decision Support to Support Protocols
• Pop up alert
• Provider Checklist
• SideBar Dashboard
31
Pop up Alert
32
Provider Checklist
33
Transparent Analytics
Data/Analytics
34
Supportive Care Metric Dashboard
35
3
ACP Note Completion
4
68
57
92
250
470
428
623
449
68
208
263
268
380
531
600
808
630
0
200
400
600
800
1,000
1,200
1,400
1,600
Qtr 3, 2016 Qtr 4, 2016 Qtr 1, 2017 Qtr2, 2017 Qtr 3, 2017 Qtr 4, 2017 Qtr 1, 2018 Qtr 2, 2018 Qtr 3, 2018
ACP Note Completion September 2016 through August 2018 (N = 6197)
Ambulatory Visit Hospital Encounter
36
3
eMOLST Completion
3
1
6
12
22
23
37
57
35
33
41
65 65
79
67
81
96
106
93
136
150
170
223
187
224
0
50
100
150
200
250
eMOLST Completion August 2016 through August 2018 (N = 2012)
37
3
eMOLST Rate for DNR Patients
38
“I would not be
surprised if this
patient passed away
in the next 6
months”
Hospice
Expired
+
+
EOL Cohort:
Compared one year of adult
inpatient activity for EOL cohort
against entire patient population
39
MSQ RESPONSE RATE
ADVANCE CARE PLANNING NOTES
MOLST DOCUMENTATION
34%
92%
92%
EOL Cohort
40
Results
– Average Daily Census (-6%,-13%)
– Readmission Rate (-3%, -5%)
– # of Hospital Acquired Conditions in Cohort (-42%, -57%)
– Total Inpatient Days (-6%, -12%)
– IP Discharges to Hospice (3%, 21%)
– Variable Direct Cost (-17%, -9%)
FY18 vs FY17 Average for Patients in EOL Cohort (Manhattan and
Brooklyn)
41
Barriers
• Integration of eMolst
• Education/Comfort
• Accountability
• Priority Fatigue
42
Further Research
• $7.5 million NIH Grant
• NYU Langone Health Ronald O. Perelman Department
of Emergency Medicine (PI: Corita Grudzen MD)
• 35 clinical sites from 18 health systems across US
43
Acknowledgments
• Leadership Team
– Nader Mherabi CIO
– Bob Press MD PhD
– Kim Glassman RN PhD
• MCIT Build Team
– Vicky Javier RN
– Dave Randhawa
– Meg Ferrauiola
– Lani Albania RN
• Value Based Management
– Nicole Adler MD
– Frank Volpicelli MD
– Steve Chatfield
– Will Winfree
• Advance Care Planning Program
– Christine Wilkins PHD LCSW
– Tom Sedgwick LCSW
44
45
• Jonathan Austrian MD Jonathan.Austrian@nyulangone.org
• Glenn Doty RN Glenn.Doty@nyulangone.org
Questions
